Pseudo Label Refinery for Unsupervised Domain Adaptation on Cross-dataset 3D Object Detection

arXiv:2404.19384v1 Announce Type: new
Abstract: Recent self-training techniques have shown notable improvements in unsupervised domain adaptation for 3D object detection (3D UDA). These techniques typically select pseudo labels, i.e., 3D boxes, to supervise models for the target domain. However, this selection process inevitably introduces unreliable 3D boxes, in which 3D points cannot be definitively assigned as foreground or background. Previous techniques mitigate this by reweighting these boxes as pseudo labels, but these boxes can still poison the training process. To …

3d object 3d object detection arxiv cs.ai cs.cv dataset detection domain domain adaptation object type unsupervised